Aging, Sleep Disturbance and Disease Status: Cross-Sectional Analysis of the Relationships Between Sleep and Multimorbidity Across the Lifespan in a Large-Scale United States Sample

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear authors, 

Comments for author File: Comments.pdf

Author Response

We'd like to thank the reviewer for their valuable suggestions that led to an improvement of the paper. We have made our revisions and are providing our responses below, thank you again for taking the time to review.  

Introduction

Comments 1: “Absence of Discussion on Existing Interventions: Current interventions for sleep disorders or multimorbidity are not addressed, limiting the contextual framework of the study.”

Response 1: Thank you so much for pointing out this area of our paper that could be improved. Due to the recent nature of published manuscripts exploring poor sleep and multimorbidity, there is not currently any existing studies exploring the impact of psychological sleep interventions on multimorbidity. Current research in this area is primarily focused upon establishing the relationship between these two factors across populations to emphasize the potential of such interventions to improve multimorbidity. Given the high incidence of both sleep disturbances and multimorbidity in the United States, the aim of this study was to establish initial evidence investigating the relationship of subjective sleep disturbances on multimorbidity status within the US. We agree that the manuscript as it stands does not explain this detail, and thus, we have added the following change, “Given that initial research exploring the impact of sleep patterns on the prevalence of multimorbidity is still in its infancy, to the author’s knowledge, there are currently no randomized controlled trials (RCTs) investigating the impact of targeted sleep interventions, such as CBT-I, on the incidence or progression of multimorbidity. Authors hope that results of this study may contribute to the robust body of evidence linking poor sleep with multimorbidity across populations and inform future research regarding the utility of psychological sleep interventions on improving quality of life in individuals with MCCs.” The revised manuscript change can be found in red on page 2, paragraph 3, line numbers 77-85.

 

Comments 2: “Lack of Detail on Vulnerable Populations: Although minoritized populations are mentioned, specific factors contributing to health disparities are not explored in depth.”

Response 2: Thank you for pointing this out. We agree that specific factors contributed to health disparities are not explored in depth. This was an intentional choice from the authors during the writing process due to the limited demographic variability observed in MIDUS’s sample. Despite small N’s for individuals within vulnerable populations within this sample, we still thought it relevant to mention disparities briefly as justification for investigating the role of race, age, and economic status as dichotomous variables to account for sample size differences across categories. To provide more context as to why this decision was made, we have added the following to the Limitations section of the manuscript, “Further, this study’s sample was predominantly socially and economically privileged sample (white, college-educated, high-income, low depressive symptoms), limiting the generalizability of study findings across diverse populations. It is unclear if this is simply a reflection of the demographic composition of the local population in which the study was conducted, or if it is indicative of recruitment bias from the original MIDUS study’s sampling procedures. Regardless, due to the underrepresentation of individuals with vulnerability identities, study findings should be interpreted with caution for individuals with intersectional or minoritized identities. This could be addressed through inclusion of targeted recruitment strategies that aim to intentionally engage with underrepresented communities to promote sufficient representation across population sub-groups.” The revised manuscript change can be found in red on page 10, paragraph 3, and line numbers 430-439.

Comments 3: “Generalization of Global Data: Studies from other countries are cited, but there is no analysis of how cultural or social differences affect the applicability of these findings to the U.S. context.”

Response 3: Thank you for pointing this out. We agree with this comment. Therefore, we have added the following discussion to the manuscript: “Due to differences in healthcare policy and infrastructure, resource access, socioeconomic disparities, and cultural norms surrounding medicine, psychology, and sleep, it is imperative that this relationship be investigated within the US to explore how these differences may uniquely impact the relationship between sleep and MCCs”. The revised manuscript change can be found in red on page 3, paragraph 1, and line numbers 101-105.

 

Comments 4: “Lack of Discussion on Modifiable Factors: Practical interventions to improve sleep and reduce multimorbidity are not explored.”

Response 4: Thank you for pointing this out. We believe this was addressed with the response provided to comment 1, in that there are no existing practical interventions to explore the relationship between both factors in conjunction. The revised manuscript change can be found in red on page 2, paragraph 3, line numbers 77-85.

 

Comments 5:  - “Absence of Justification for Focusing on Sleep: There is no explanation as to why sleep disorders were prioritized over other risk factors.”

Response 5: Thank you for pointing this out. We respectfully note that the rationale for prioritizing sleep disturbance is provided in detail in Section 1.2, Sleep Disturbances, Chronic Illness, Healthcare Utilization, and Quality of Life. This section outlines the high prevalence of sleep disturbance, its bidirectional relationship with chronic conditions, its economic burden, and its potential role as a modifiable risk factor. Additionally, the literature review highlights the growing body of global research in this area and the need to examine these patterns specifically within the U.S. context. Therefore, we believe the justification for the study’s focus on sleep is adequately supported. We are happy to address any specific concerns that may be raised, but we opted respectfully not to change this section based upon the above comment.

 

Methods

 

Comments 6: “Limited Definition of Multimorbidity: The operationalization of multimorbidity may vary, making it difficult to generalize findings across different healthcare contexts.”

Response 6: Thank you for pointing this out. We agree with this comment and have discuss the challenges with assessing multimorbidity in the limitation settings, as well as discussed, its limited generalizability. Given that this analysis was completed on data from a secondary database, the authors of this study were not involved in the original MIDUS study design or data collection process. Multimorbidity, as conceptualized within this study, has been quantified similarly in previous empirical research, but this was not explicitly stated within the manuscript. Therefore, we have added the following to the manuscript: “Notably, one study, emanating in Australia, supports an operationalization of multimorbidity as a count variable, representing the number of chronic conditions when two or more disease states are present, similar to the quantification of multimorbidity within this analysis [76]. Despite this, other research demonstrating the equivalence of multimorbidity assessment has reported mixed findings [76–78], indicating it is possible that the results of this study may have varied depending on which quantification of multimorbidity was used”. This change can be found on page 10, paragraph 2, and line numbers 400-406.

 

Comments 7: “Subjective Nature of Assessments: The use of questionnaires and telephone interviews may introduce social desirability bias and compromise data validity.”

Response 7: Thank you so much for pointing this out! This study does utilize subjective assessments due to the availability of data collected through MIDUS. This study sought to determine an initial relationship between subjective sleep disturbances and multimorbidity to establish the connection between these factors within the United States. The subjective nature of assessments as well as the potential for social desirability biases was discussed in the limitations section. Further, it was mentioned in the manuscript that future studies should seek to address these biases through the inclusion of objective sleep assessment. Current authors hope to address this by repeating analyses in the future with the inclusion of objective sleep data. Therefore, no changes were made to the manuscript.

 

Comments 8: “Lack of Objective Sleep Measures: Only subjective sleep measures, such as the PSQI, were used, limiting the precision of the findings.”

Response 8: Thank you for pointing this out. We agree that this limits the precisions of study findings, as discussed in both the limitations and future directions sections of the paper. Due to this analysis being conducted upon secondary data, researchers were limited in choices regarding available sleep data collected previously from participants. However, the PSQI has empirical support demonstrating both its validity and reliability across multiple populations, as mentioned in the manuscript. Authors hope to conduct follow-up studies in the future that address this limitation through inclusion of objective sleep metrics to assess its impact on chronic health conditions. Given that these limitations were previously discussed in the study body, no further changes were made.

 

Comments 9: “Predominantly Privileged Sample: The sample, composed mostly of White, highly educated, and high-income individuals, limits generalizability to more diverse or underserved populations.”

Response 9: Thank you for pointing this out. We agree with this comment, and believe it was addressed with the changes made in response to comment 1 (Introduction). This change discusses the limited diversity within study sample, as well as how it limits applicability of study findings across populations. The revised manuscript change can be found in red on page 2, paragraph 3, line numbers 77-85.

 

Comments 10: “Absence of Longitudinal Data: The cross-sectional design prevents the analysis of how sleep disturbances and multimorbidity evolve over time.”

Response 10: Thank you for pointing this out. We agree with this comment. Due to the cross-sectional nature of data collected, it is not possible to assess how this relationship persists over time. The manuscript discusses this facet within the limitations section, as well as recommended options to address the lack of temporal analysis. Authors hope to address how time may impact this relationship in future studies, however, authors used available data to determine an initial relationship. We are happy to provide further information on this if specific suggestions are made. Therefore, the manuscript was not changed in response to this suggestion. 

 

Results

 

Comments 11: Lack of Subgroup Analysis: There is no detailed examination of how results vary across demographic subgroups, such as racial minorities or low-income populations.

Response 11: Thank you for pointing this out. We agree that a more detailed examination of results across demographic subgroups would be informative. However, this was not feasible given the sample characteristics of the MIDUS dataset. While MIDUS collected detailed demographic data (e.g., race, income, education), the small sample sizes in certain subgroups—particularly among racially minoritized and low-income individuals—led to convergence issues in regression models, including Hessian matrix warnings. To ensure model stability, we were required to collapse categories into dichotomous variables (e.g., White vs. Non-White). Therefore, we have added a statement to the Future Directions section noting the need for future studies to ensure more balanced subgroup representation to support stratified analyses, “Finally, future research should prioritize diverse recruitment strategies with the goal of obtaining a representative sample so that future researchers may stratify analyses to better investigate the role of health disparities on multimorbidity”. The revised manuscript change can be found in red on  page 11, paragraph 2, and line numbers 462-464.

 

Comments 12:  Limited Impact of Non-significant Variables: Variables such as physical activity and marital status were not significant, yet no discussion is provided as to why.

Response 12: Thank you for pointing this out. We agree that the lack of discussion around non-significant predictors may limit interpretability. Variables such as physical activity and marital status did not emerge as significant predictors in the final model. This could be due to measurement limitations (e.g., self-reported physical activity levels with no correlative objective assessment), sample characteristics, or due to data analytic constraints in which relationship status had to be dichotomized to: currently married, not currently married. We have added the following sentence in the Discussion section to acknowledge these possibilities and highlight directions for future exploration, “Notably, results of the current study did not align with previous empirical research in chronic illness, in that subjective physical activity levels and current marital status were not reported to be significant predictors for multimorbidity within this analysis. While the precise cause of this is unknown, potential factors influencing this result may have been the homogeneity of the sample, or limitations inherent to the structure and style of included assessments (ex. yes/no questions, self-report style). Future research should not discredit the importance of these variables upon the development of multiple chronic health conditions until their relative contributions are assessed without these measurement constraints.” The revised manuscript change can be found in red on page 9, paragraph 2, and line numbers 350-358.

 

Comments 13:  Lack of Exploration of Interactions: No analyses were conducted on interactions between variables, such as the combined effect of depression and sleep disorders on multimorbidity.

Response 13: Thank you for this helpful suggestion. We agree that exploring interactions—particularly between psychological and behavioral variables—could provide deeper insight into how multimorbidity develops. Due to limitations in statistical power and model convergence issues, we prioritized testing main effects in this study. However, we agree that future studies should assess whether variables such as depression and sleep disturbance interact to influence multimorbidity risk. A statement reflecting this has been added to the Future Directions section of the manuscript, “While this study primarily explored main effects, future research may also consider investigation of interaction amongst predictor variables to tailor interventions for individuals at higher risk for multimorbidity”. The revised manuscript change can be found in red on page 11, paragraph 2, and line numbers 456-458.

 

Discussion

 

Comments 14: Lack of Explanation for Non-significant Results: The discussion does not address why variables such as physical activity and marital status were not significant predictors.

 

Response 14: Thank you so much for pointing this out! We believe this comment was

addressed through our response to comment 12 in the above section. Therefore, no

further changes were made following this comment. The previously revised manuscript change can be found in red on page 9, paragraph 2, and line numbers 350-358.

 

Comments 15: Absence of Comparison with Global Studies: Although international studies are mentioned, there is no direct comparison of the findings.

Response 15: Thank you for pointing this out. We agree that explicitly comparing our findings with those of global studies enhances the contextual value of this research. To address this, we have added a comparison to the Discussion section highlighting similarities and differences between findings from international samples and our US-based sample. The following section was added/modified, “Interestingly, despite differences in international healthcare infrastructure, resource access, and cultural norms surrounding medicine, findings from this preliminary US sample reflected similar patterns between disturbed sleep and multimorbidity as global studies based in Brazil, China, and Sweden [26, 42-44]. Implications of this research within the US may suggest the integration of assessment of relevant variables that predict multimorbidity (sleep disturbances, subjective health rating, depression level, and risk for SDOH) within the primary care setting. This may be an additional strategy to help attenuate risk for MCC’s”. The revised manuscript change can be found in red on page 10, paragraph 1, and line numbers 385-392.

 

Comments 16: Lack of Exploration of Causal Factors: The causal mechanisms linking sleep disorders to multimorbidity are not investigated.

Response 16: Thank you for this insightful comment. We agree that identifying causal mechanisms underlying the relationship between sleep disturbance and multimorbidity is a critical area for future research. However, given the cross-sectional design of this study and the limitations inherent to secondary data analysis, establishing causality was beyond the scope of the current investigation. While we discussed associations between sleep and chronic illness, we intentionally avoided making causal inferences. As such, no changes were made to the manuscript in response to this comment, though we hope the study contributes foundational evidence that may inform longitudinal and mechanistic work in the future.

 

Comments 17: Limited Generalization: The discussion does not address how the findings could be applied to diverse or underserved populations.

Response 17: Thank you for pointing this out. We agree that generalization to diverse or underserved populations is an important consideration. However, due to the demographic makeup of the MIDUS sample and limitations in subgroup representation, we were not able to explore this empirically. As such, we have added text to the Limitations section emphasizing that the applicability of findings to underserved groups is uncertain and that future research should intentionally recruit more representative samples to assess these relationships across sociodemographic strata. The revised manuscript change can be found in red on page 11, paragraph 1, and line numbers 430-439.

 

Comments 18: Lack of Intervention Proposals: No practical interventions based on the study’s findings are suggested.

Response 18: Thank you for this suggestion. While we agree that practical intervention development is a critical future step, the field is still in the preliminary stages of establishing the relationship between sleep disturbance and multimorbidity. As such, no current interventions specifically target multimorbidity via sleep treatment. We have clarified this point in the Discussion section and noted that our study contributes to the foundational evidence needed to eventually inform intervention development via the following, “At the current moment, no such interventions exist that target multimorbidity risk through sleep, however, this study contributes valuable foundational evidence that can be used to guide eventual intervention development.” The revised manuscript change can be found in red on page 9, paragraph 1, and line numbers 346-349.

 

Conclusion

Comments 19: Lack of Detail on Applicability: The study does not explore how the findings could be translated into clinical practice or public policy.

Response 19: Thank you for this thoughtful comment. We agree that emphasizing how findings may inform clinical, or policy directions is important. As this study was designed to explore an initial association between subjective sleep disturbance and multimorbidity, its direct application to clinical practice or policy is limited at this stage. However, we believe these findings may support increased attention to sleep disturbance in primary care settings and public health messaging related to chronic disease prevention. Therefore, we have added a brief statement in the conclusion section outlining this potential, “Whilst findings were exploratory, study results emphasize the potential utility of screening for sleep disturbances in patients with MCC’s to identify sleep as a modifiable risk factor for disease”. The revised manuscript change can be found in red on page 11, paragraph 3, and line numbers 470-472.

 

 Comments 20: Absence of Future Research Proposals: No suggestions are made for future studies to address the identified gaps.

Response 20: Thank you for this comment. We respectfully note that several future research directions are already outlined in the manuscript. These include suggestions to utilize longitudinal designs, incorporate objective sleep assessments, explore subgroup differences using more representative samples, and investigate potential interaction effects (e.g., between sleep disturbance and depression). As these proposals are integrated throughout the Discussion and Future Directions sections, no additional changes were made to the manuscript in response to this comment.

 

Comments 21: Narrow Focus on Sleep: Although sleep is emphasized as an important factor, other potentially relevant factors for multimorbidity are not considered.

Response 21: Thank you for this feedback. We agree that multimorbidity is multifaceted and influenced by a range of behavioral, psychological, and social factors. However, this study intentionally focused on sleep due to its growing relevance in the chronic disease literature and its underrepresentation in multimorbidity research. To clarify this scope, we have modified the last section of the introduction to explicitly clarify that sleep disturbance and multimorbidity are the primary variables of interest within this manuscript. No changes were made to other sections of the manuscript.

 Comments 22: Lack of Discussion on Impact for Vulnerable Populations: The potential benefits of interventions for minoritized or low-income populations are not addressed.

Response 22: Thank you for this comment. As noted in our response to Comment 18, there are currently no sleep-focused interventions developed specifically to address multimorbidity outcomes. As such, we did not speculate on intervention effects. However, we agree that any future interventions could hold particular relevance for minoritized and low-income populations given the disproportionate burden of both sleep disturbance and multimorbidity in these groups. We have chosen not to make a change to the manuscript at this time, given the early-stage nature of this research and the lack of existing intervention data

Reviewer 2 Report

Comments and Suggestions for Authors

The present manuscript reports relationships between sleep and multimorbidity across the lifespan in a large-scale United States sample. Investigation of factors associated with two or more chronic medical conditions has a potential to identify vulnerability markers, mitigate negative health consequences and promote longevity. However, several critical issues are problematic in the manuscript that distract from the overall effort.

Several concerns are mentioned below:

 

• Title - “….Disease acquisition…” The manuscript does not provide sufficient data to talk about disease acquisition; only cross-sectional data are analyzed.
• Abstract- should be modified after reviewers’ comments will be addressed
• Line 96-97 - The study aim is not clear - with no information regarding sleep disturbances in this population in the past 12 month (sleep disturbances were present or newly developed), one is unable to determine the actual impact of sleep disturbances in relation to chronic medical conditions.
• Methodology - Lines 100-114 - needs better description; What are project 1 and project 4? Were there projects 2 and 3? What biomarkers? If MIDUS 2 and MIDUS 3 employed same measures, why longitudinal data are not present?
• Line 119 - “Participants were asked if they had experienced symptoms of 30 chronic health conditions within the past 12 months” - symptoms of chronic condition in the past does not necessarily indicate that subject actually has a chronic condition. More detailed information about how chronic conditions were determined is needed.
• Methods - A separate paragraph describing how data were analyzed, including statistical tests, is missing. Subsection 3.3. Model Fit should be part of “Data Analyses”.
• Results - I recommend clarifying the use and interpretation of the incidence rate ratio (IRR) in the manuscript, since it lacks traditional meaning of emerging new cases
• Results - In my opinion, chronic medical conditions in the past 12 month and sleep disturbances assessed in the past month makes discussion about prediction and/or incidence rates questionable. It seems impossible for current sleep problems (or covariates assessed in the study) to truly predict health conditions that were present in the past 12 months, although they are truly associated.
• Discussion - Again, in my opinion, “Findings from this study contribute to previous literature demonstrating that chronic health conditions are multifactorial and can be predicted by various demographic, physiological, and psychological variables” - with your data you cannot claim prediction. This is a critical impediment to a clear interpretation of the findings presented.
• Discussion should be rewritten upon suggested changes within the manuscript.

 

 

Author Response

   We'd like to thank the reviewer for their valuable suggestions that led to an improvement of the paper. We have made our revisions and are providing our responses below, thank you again for taking the time to review.  

Comments 1: Title - “….Disease acquisition…” The manuscript does not provide sufficient data to talk about disease acquisition; only cross-sectional data are analyzed.

Response 1: Thank you for pointing this out. We agree with this comment. Therefore, we have changed the title from ‘Disease acquisition’ to ‘Disease status’. The revised manuscript change can be found in red on the page 1, paragraph 1, line 2.

Comments 2: Abstract- should be modified after reviewers’ comments will be addressed

Response 2: Thank you so much for pointing this out. We have revised the abstract after addressing comments from each reviewer. The adjusted abstract can be found on page 1, paragraph 1, lines 11-25.

Comments 3: Line 96-97 - The study aim is not clear - with no information regarding sleep disturbances in this population in the past 12 months (sleep disturbances were present or newly developed), one is unable to determine the actual impact of sleep disturbances in relation to chronic medical conditions.

Response 3: Thank you for this comment. We agree that the lack of longitudinal data limits the ability to draw conclusions about the temporal relationship between sleep disturbances and chronic health conditions. However, the primary aim of this study was to examine the association between subjective sleep disturbances and multimorbidity at a single time point, using available cross-sectional data from MIDUS. As such, we did not seek to assess the onset or chronicity of sleep issues, but rather whether poor sleep quality was associated with greater multimorbidity burden at the time of assessment. This has been acknowledged as a limitation in the manuscript. We have clarified the language within the study to emphasize the correlational approach to our analysis. The revised manuscript change can be found at page 10, paragraph 3, line numbers 424-429.

Comments 4: Methodology - Lines 100-114 - needs better description; What are project 1 and project 4? Were there projects 2 and 3? What biomarkers? If MIDUS 2 and MIDUS 3 employed same measures, why longitudinal data are not present?

Response 4: Thank you for these clarifying questions. Project 1 within MIDUS-2 refers to the main longitudinal survey follow-up of the original MIDUS cohort, and Project 4 refers to the biomarker subproject, which included in-person biological assessments for a subset of participants. Projects 2 and 3 involved neuroscience and twin study components, respectively, but were not relevant to the variables analyzed in the current study. Therefore, they were not described in detail. As noted in the manuscript, only MIDUS-2 data were used for the current analysis due to the lack of comparable biomarker data in MIDUS-1 and the timing of variable collection. Although MIDUS-3 includes overlapping measures, this study was designed as a cross-sectional analysis using the MIDUS-2 wave specifically. Additional detail on MIDUS study structure and project descriptions can be accessed through the publicly available MIDUS documentation at https://www.icpsr.umich.edu/web/NACDA/studies/4652. We agree that this was not clear in the manuscript, so we added the following sentence to address this lapse in clarity, “Alternate projects involving the collection of neuroscience and twin study data (Projects 2 & 3), are not relevant to the current analysis and therefore, are not described in full here. Further information on all MIDUS projects is available via MIDUS’s public data repository [48].” The revised manuscript change can be found in red on page 3, paragraph 2, line numbers 128-131.

Comments 5:  Line 119 - “Participants were asked if they had experienced symptoms of 30 chronic health conditions within the past 12 months” - symptoms of chronic condition in the past does not necessarily indicate that subject actually has a chronic condition. More detailed information about how chronic conditions were determined is needed.

Response 5:  Thank you for your observation. We agree that having chronic conditions is not equivalent to experiencing symptoms. The presence of chronic conditions are self-reported in the MIDUS study and the language has been altered accordingly. Examples of prevalent conditions have been listed in section 2.2 Measures, Multimorbidity to make a clear distinction between health symptoms and reported chronic conditions. The revised manuscript change can be found in red on page 4, paragraph 1, line numbers 143-145.

Comments 6: Methods - A separate paragraph describing how data were analyzed, including statistical tests, is missing. Subsection 3.3. Model Fit should be part of “Data Analyses”.

Response 6: Thank you so much for pointing this out! Following this recommendation, we have added an additional section entitled, 3.3 Data Analyses, and moved information surrounding Model Fit into this subsection. The revised manuscript change can be found in red on page 5, paragraph 2, line numbers 206-226.

Comments 7: Results - I recommend clarifying the use and interpretation of the incidence rate ratio (IRR) in the manuscript, since it lacks traditional meaning of emerging new cases

Response 7: Thank you for this comment. Per our understanding of the interpretation of negative binomial regression outputs, the variable of interest, Exp(B), is the incidence rate ratio (IRR). It is the exponentiated value of the original regression coefficient, reflected from the logarithmic scale, back to the original outcome scale (Hennigan, 2021). The following sentence was added to the manuscript to clarify interpretation of the model, “The NBR model was interpreted through examining the incidence rate ratio (IRR), which reflects the exponentiated value of the original regression coefficient, reflected in the original scale of the DV [87]. The IRR variable quantifies a predictive effect through reporting the expected multiplicative change in the dependent variable per each one-unit increase in each predictor variable, whilst holding other variables constant.” The revised manuscript change can be found in red on page 6, paragraph 1, and line numbers 250-255.

Comments 8: Results - In my opinion, chronic medical conditions in the past 12 months and sleep disturbances assessed in the past month makes discussion about prediction and/or incidence rates questionable. It seems impossible for current sleep problems (or covariates assessed in the study) to truly predict health conditions that were present in the past 12 months, although they are truly associated.

Response 8: Thank you for this thoughtful comment regarding the temporal mismatch between the assessment periods for sleep disturbances (past month) and chronic medical conditions (past 12 months). We agree that this discrepancy limits the ability to draw conclusions about temporal prediction or incidence. As noted, the cross-sectional nature of the data and the lack of information on the onset timing of either sleep disturbances or chronic conditions preclude causal conclusions.

However, it is important to note that this methodological approach—assessing recent sleep disturbances in relation to chronic health conditions within the past year—is consistent with prior research in this area and reflects common practice in modeling associations between sleep and chronic disease outcomes (Idalino et. Al, 2023; Calderón-Larrañaga et. al, 2021; Sindi et. al, 2020). Although temporality cannot be confirmed, sleep disturbances are often conceptualized as a potential risk indicator or correlate of chronic disease burden in similar analytic frameworks.

While we use the term “predict” in the statistical sense—aligned with conventions in Negative Binomial Regression modeling—respectfully, we have clarified in the manuscript that the goal of this analysis is to examine correlational associations, not to infer causality (Hennigan, 2021). However, we appreciate this comment as it points to our lack of clarity of explanation in the manuscript, and we have ensured that limitation has been explicitly acknowledged in the revised limitations section to support appropriate interpretation of the findings.

Comments 9: Discussion - Again, in my opinion, “Findings from this study contribute to previous literature demonstrating that chronic health conditions are multifactorial and can be predicted by various demographic, physiological, and psychological variables” - with your data you cannot claim prediction. This is a critical impediment to a clear interpretation of the findings presented.  Discussion should be rewritten upon suggested changes within the manuscript.

Response 9:  Thank you very much for this thoughtful and constructive comment. We recognize the importance of clearly distinguishing between statistical associations and true predictive or causal relationships, particularly when working with cross-sectional data. We appreciate your observation that the use of the term “predict” may unintentionally imply a stronger temporal or causal interpretation than is warranted by our study design and intended to only utilize this term in the statistical sense.

In response, we have carefully revised the Discussion section to reflect this concern. Specifically, we have adjusted language to emphasize that our findings highlight associative relationships between demographic, physiological, and psychological factors and chronic health conditions, rather than making claims of prediction. The sentence in question has been revised to read: “Findings from this study contribute to previous literature demonstrating that chronic health conditions are multifactorial and are associated with various demographic, physiological, and psychological variables.” The revised manuscript change can be found in red on page 8, paragraph 1, and line numbers 313-315.

We have also reviewed the manuscript to ensure consistency in wording and to avoid any overstatement of the data’s implications. We hope these revisions provide a clearer and more accurate interpretation of the study’s contributions and limitations.

References

Calderón-Larrañaga, A., Pérez, L., Vetrano, D., Triolo, F., Sjöberg, L., Darin-Mattsson, A., Inzitari, M., & Sindi, S. (2021). Sleep Disturbances and the Speed of Multimorbidity Development in Old Age. Innovation in Aging, 5(Supplement_1), 377. https://doi.org/10.1093/geroni/igab046.1462

Hennigan, P. (2021). Negative Binomial Regression Guide. https://doi.org/10.13140/RG.2.2.24396.74885/1

Idalino, S. C. C., Canever, J. B., Cândido, L. M., Wagner, K. J. P., de Souza Moreira, B., Danielewicz, A. L., & de Avelar, N. C. P. (2023). Association between sleep problems and multimorbidity patterns in older adults. BMC Public Health, 23(1), 978. https://doi.org/10.1186/s12889-023-15965-5

Sindi, S., Pérez, L. M., Vetrano, D. L., Triolo, F., Kåreholt, I., Sjöberg, L., Darin-Mattsson, A., Kivipelto, M., Inzitari, M., & Calderón-Larrañaga, A. (2020). Sleep disturbances and the speed of multimorbidity development in old age: Results from a longitudinal population-based study. BMC Medicine, 18(1), 382. https://doi.org/10.1186/s12916-020-01846-w

 

Reviewer 3 Report

Comments and Suggestions for Authors

Dear authors,

Thank you for your submission and contribution.

Great study! This study is timely and aligns well with sustainable development goal related to good health and well-being.

Overall, the paper is well-written and informative. I have few review comments, recommendations, and suggestions for your consideration to enhance paper overall readability. 

Abstract 

For participants, please add more demographic information and descriptive data (sample characteristics: i.e., age, gender, diagnoses, etc.) while still adhering to abstract word count limit. 

Keywords: it is recommended to arrange keywords in an alphabetical order. 

Materials and Methods
Participant: please add recruitment method and inclusion and exclusion criteria

how did you determine sample size? were sample size calculations performed? please explain

self-report questionnaires used in this study: who did develop them? were they pilot-tested and validated for the purpose of this study? could you provide a brief of content and description of questions included?  

I will be happy to look at the revised version. 

Best wishes, 

Author Response

     We'd like to thank the reviewer for their valuable suggestions that led to an improvement of the paper. We have made our revisions and are providing our responses below, thank you again for taking the time to review.  

Abstract

1. Comments 1: For participants, please add more demographic information and descriptive data (sample characteristics: i.e., age, gender, diagnoses, etc.) while still adhering to abstract word count limit. 

Response 1: Thank you for pointing this out. We agree with this comment. Therefore, we have added some of the relevant demographic information and descriptive data to the abstract. This change can be found on page 1, abstract, lines 17-19.

 

2. Comments 2: It is recommended to arrange keywords in an alphabetical order. 

Response 2: Thank you for pointing this out. We agree with this comment. Therefore, we have rearranged the keywords in alphabetical order. This change can be found on page 1, in the keywords section, line 25.

 

Materials and Methods

3. Comments 3: Please add recruitment method and inclusion and exclusion criteria.

Response 3: Thank you for pointing this out. All recruitment, eligibility, and inclusion/exclusion were done at the time of original study data collection and was independent of the current analysis. However, we agree that more information should be provided about the original study’s methodology. Therefore, we have added further information about the study’s original eligibility criteria, as well as provided the resource where the original study’s full methodology can be accessed. This change can be found on page 3, paragraph 3, lines 120-125.

 

4. Comments 4: How did you determine sample size? Were sample size calculations performed? Please explain.

Response 4: Thank you for pointing out a lack of discussion regarding sample size. We followed guidelines suggested by Green (1991) regarding the procedure used to determine an adequate sample size for regression, in which N>50+8m (m= number of IV’s) is used [1]. Per this calculation, given that there were 11 independent, predictor variables, a minimum sample size of (N=138 participants) would be required to examine relationships. Given that the number of participants within our study far exceeded this value (N=1053 participants), we performed no further sample size calculations. We have not added this to the manuscript but are happy to make the change if further warranted.

 

5. Comments 5: self-report questionnaires used in this study: who did develop them? were they pilot-tested and validated for the purpose of this study? could you provide a brief of content and description of questions included?

Response 5: Thank you for pointing out the lack of clarification regarding the self-report questionnaires used in this study. All self-report questionnaires used in this analysis were detailed in the proposed manuscript. Additional information gleaned during phone-interviews were designed by the MacArthur Midlife Research Network for the purpose of collecting demographic and lifestyle information from participants. They were not pilot-tested or validated for the purpose of the study. A full account for interview questions can be found under the DOI listed on page 3, paragraph 3, lines 120-125.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Introduction

Lack of randomized controlled trials (RCTs): There are no RCTs investigating the impact of sleep-targeted interventions, such as Cognitive Behavioral Therapy for Insomnia (CBT-I), on the incidence or progression of multimorbidity.

Method

Subjective nature of measures: The use of telephone interviews and questionnaires may introduce social desirability bias and limit data accuracy.
Temporal inconsistency: Chronic conditions were assessed over a 12-month period, whereas sleep disorders were evaluated over a 1-month period, hindering longitudinal analysis.

Results

Non-significant variables: Physical activity and marital status were not significant predictors of multimorbidity, which contradicts previous findings. This may be due to sample homogeneity or limitations in assessment methods.

Discussion

Lack of interaction effects: The study focused on main effects without exploring potential interactions between variables, which could provide more nuanced insights.

Subjective sleep measures: The reliability of subjective sleep measures, such as the Pittsburgh Sleep Quality Index (PSQI), is questionable compared to objective methods like polysomnography.

Conclusion

Lack of targeted interventions: There are no existing interventions addressing the risk of multimorbidity through sleep disorders, despite the evidence presented.

Author Response

Comments 1: Lack of randomized controlled trials (RCTs): There are no RCTs investigating the impact of sleep-targeted interventions, such as Cognitive Behavioral Therapy for Insomnia (CBT-I), on the incidence or progression of multimorbidity.

Response 1: Thank you for reiterating this important point. We agree that the absence of randomized controlled trials (RCTs) examining the impact of sleep-targeted interventions, such as Cognitive Behavioral Therapy for Insomnia (CBT-I), on the incidence or progression of multimorbidity represents a notable gap in the literature. This issue was addressed in our previous revision in response to an earlier comment regarding the lack of intervention proposals (Comment 18).

Specifically, we have clarified in the Discussion section that while no known RCTs currently examine sleep-based interventions for the prevention or treatment of multimorbidity, this study contributes foundational evidence that may support future intervention development. We also noted that the field is still in the early stages of establishing this association. We hope this addition adequately reflects the current state of the literature and the need for future clinical trials in this area. We remain grateful for your attention to this important point.

 

Comments 2: Subjective nature of measures: The use of telephone interviews and questionnaires may introduce social desirability bias and limit data accuracy.

Response 2: Thank you for highlighting this important methodological consideration. We agree that reliance on self-reported measures via telephone interviews and questionnaires introduces the possibility of social desirability bias and may limit the precision of the data. This concern aligns closely with a previous comment (Comment 8), which we addressed by noting this limitation explicitly in both the Limitations and Future Directions sections of the manuscript.

As described, the use of subjective sleep measures (e.g., PSQI) reflects the constraints of working with secondary data, where the inclusion of objective assessments was not possible. However, we also emphasized that the PSQI is a widely used, validated instrument with strong psychometric properties across diverse populations.

We have retained our prior revisions and respectfully note that these issues are currently acknowledged in the manuscript. Should further clarification be helpful, we would be happy to elaborate further in the relevant sections.

Comments 3: Temporal inconsistency: Chronic conditions were assessed over a 12-month period, whereas sleep disorders were evaluated over a 1-month period, hindering longitudinal analysis.

Response 3: Thank you for raising the issue of differing timeframes for chronic condition and sleep assessments. We fully acknowledge that evaluating chronic conditions over 12 months alongside 1‑month sleep data limits meaningful longitudinal conclusions. This temporal inconsistency aligns with concerns previously raised (e.g., Comment 10 on absence of longitudinal data).

As noted in our questionnaire and analysis limitations, the cross-sectional nature of this study precludes examination of how temporal patterns evolve. This limitation is discussed in the Limitations section, along with a recommendation for future longitudinal studies.
No further revisions were made, but the concern is fully acknowledged in the manuscript.

 

Comments 4: Non-significant variables: Physical activity and marital status were not significant predictors of multimorbidity, which contradicts previous findings. This may be due to sample homogeneity or limitations in assessment methods.

Response 4: Thank you for this insightful point. We agree that the non-significance of physical activity and marital status—contrary to prior literature—warrants discussion.

This issue is addressed in our Response 12 (Comment 12), where we added a nuanced discussion in the Discussion section. We note that measurement limitations (e.g. self-reported physical activity; dichotomized marital status) and sample homogeneity likely influenced these findings.
The revised text includes these considerations and is located in red on page 9, paragraph 2 (lines 350–358). No additional edits were made since the topic is comprehensively covered.

Comments 5: Lack of interaction effects: The study focused on main effects without exploring potential interactions between variables, which could provide more nuanced insights.

Response 5: Thank you for this helpful suggestion. We agree that testing interaction effects—such as depression by sleep disturbance—could yield deeper insights into multimorbidity risks.

As described in Response 13 (Comment 13), we have explicitly acknowledged this in the Future Directions section. While statistical power and convergence concerns guided our focus on main effects in this study, we highlight the importance of investigating interactions in future research. The added text is located on page 11, paragraph 2 (lines 456–458). No further edits were made, as the suggestion is already woven into the manuscript.

 

Comments 6: Subjective sleep measures: The reliability of subjective sleep measures, such as the Pittsburgh Sleep Quality Index (PSQI), is questionable compared to objective methods like polysomnography.

Response 6: Thank you for emphasizing this important methodological issue. We agree that reliance on the Pittsburgh Sleep Quality Index (PSQI) and telephone-based self-report may introduce bias and limit data validity.

As previously addressed in Response 8 (Comment 8), the limitations of subjective sleep measures are discussed in both the Limitations and Future Directions sections. We explain that as secondary data analysis, researchers were constrained to existing instruments, and we reaffirm PSQI’s validated reliability across populations. We also reiterate our intent to include objective sleep metrics (e.g., actigraphy or polysomnography) in follow-up studies. No further updates were made, but the issue is again clearly acknowledged.

 

Comments 7: Lack of targeted interventions: There are no existing interventions addressing the risk of multimorbidity through sleep disorders, despite the evidence presented.

Response 7: Thank you for reiterating this important point. We feel as though this comment was addressed via comment 1.

Reviewer 2 Report

Comments and Suggestions for Authors

The revised manuscript has addressed most of comments. However, one issue remains problematic. Having symptoms of a chronic disease does not necessarily indicate that one has the disease. Moreover, symptoms of several diseases may overlap. In limitation, line 403, reference 76 – is not similar to your way of the quantification of multimorbidity – in that study data was collected from general practitioners based on whether a condition was chronic in the patient. Therefore, add to the limitations that data collection in your study was based on subjective assessment, and no information about confirmed diagnoses of chronic conditions was available.

Check the clarity - lines 78-82

Author Response

Comments 1: The revised manuscript has addressed most of comments. However, one issue remains problematic. Having symptoms of a chronic disease does not necessarily indicate that one has the disease. Moreover, symptoms of several diseases may overlap. In limitation, line 403, reference 76 – is not similar to your way of the quantification of multimorbidity – in that study data was collected from general practitioners based on whether a condition was chronic in the patient. Therefore, add to the limitations that data collection in your study was based on subjective assessment, and no information about confirmed diagnoses of chronic conditions was available.

Response 2: Thank you for this thoughtful comment! Having re-read the section, we agree with your comment and have made the suggested change to the limitations section, “While the quantification of multimorbidity in this study was similar to our own (counting the number of chronic conditions), the current study relied solely on subjective assessment of chronic disease symptoms. Thus, the current study was limited due to a lack of objective CHC data as no confirmatory information regarding diagnosed medical conditions was provided. Other research presented mixed findings regarding the equivalence of multimorbidity assessment [76–78], indicating the possibility that current study results may have varied depending on which quantification of multimorbidity was used.”

 

Comments 2: Check the clarity - lines 78-82

Response 2: Thank you for this comment! We agree the sentence was originally unclear. We have corrected the line to read, “Given that initial research exploring the impact of sleep patterns on the prevalence of multimorbidity is still in its infancy, no randomized controlled trials (RCTs) have investigated the impact of targeted sleep interventions, such as CBT-I, on the incidence or progression of multimorbidity yet.” Please see the following change on page 2, paragraph 2, lines 78-82.

Reviewer 3 Report

Comments and Suggestions for Authors

Dear authors,

Thank you for the clarification and for addressing all review comments. 

Best wishes, 

Author Response

Comments 1: Dear authors, Thank you for the clarification and for addressing all review comments. Best wishes, 

Response 1: Thank you so much for all of your helpful feedback! We appreciate the time and effort you've put into helping us improve our manuscript!